Discussion Forum

Hai,
I have been trying to simulate the model of supercapacitor. Based on the li_battery model from model library , I gave all inputs required for my model to simulate. I hope I gave all my boundary conditions. While solving this model I am consistently getting the error.

Failed to find consistent initial values.
Last time step is not converged.

I have no clue how to proceed and I have tried few suggestions from this forum , those include solving stationary, changing solver. While changing solver to PARDISO, I was ending up with this error

Failed to find consistent initial values.
The_relative_error_is_greater_than_the_relative_tolerance
Relative_residual: 3e-08, Relative_error: 3.6e+02
Last time step is not converged.

my model contains following, {all derivatives are partial}

in electrode phase

d/dx(-sigma*d/dx(phi1))= -[S_d*C_d*d/dt(phi1-phi2)+S_f*jf]
BC
i_1=-sigma*d/dx(phi1)=Icell at 1
i_1 = 0 at 2
phi1= V_0= 0.5 V at t=0
faradaic reaction in subdomain expression
jf = i_0[exp(alpha_a*eta*F/(R*T))-exp(-alpha_c*eta*F/(R*T))]

in electrolyte phase
d/dx(-kp*d/dx(phi2)-kp*RT*K_junc*1/c2*d/dx(c2)/F) = S_d*C_d*d/dt(phi1-phi2)+S_f*jf

BC
i_2=-kp*d/dx(phi2)-kp*RT*K_junc*1/c2*d/dx(c2)/F = 0 at 1
i_2= I_cell at 2
phi2= -V_0= -0.5V

concentration of electrolyte gradient
d/dt(c2) = d/dt(D_s*d/dx(c2))+[S_d*C_d/(2*F*epsilon)*d/dt(phi1-phi2)+S_f*jf/(2*F*epsilon)](1-t_p0)

c2 = C_S0 at t =0
insulation at 1
flux at 2


d/dt(theta)= S_f*jf/(Qfox-Qfre)
theta = 1 at t =0



If someone could shed some light on this I would greatly appreciate it.
Here, I am attaching my model file in version 3.5a.
If you need any additional information, kindly ask for it.

thanks in advance
Ganesh. M